Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Nicolas George <george@nsup.org>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH v3] libavutil: Add AVMap
Date: Thu, 17 Apr 2025 19:57:46 +0200
Message-ID: <aAFBGsbuU5SG2ma4@phare.normalesup.org> (raw)
In-Reply-To: <20250417165241.3266643-1-michael@niedermayer.cc>

Michael Niedermayer (HE12025-04-17):
> +AVMap *av_map_new(AVMapCompareFunc cmp_keyvalue, int cmp_flags, AVMapCopyFunc copy, AVMapFreeFunc freef)
> +{
> +    AVMap *s = av_mallocz(sizeof(*s));
> +    if (!s)
> +        return NULL;
> +
> +    s->copy          = copy;
> +    s->freef         = freef;
> +
> +    av_map_add_cmp_func(s, cmp_keyvalue, cmp_flags);
> +
> +    return s;
> +}

If you do not want to do the version where dynamic allocation can be
avoided, would you at least consider using _alloc() for this? It is only
to letters more, it is consistent with the rest of the API and it would
leave _new() available if somebody adds the lighter version.

I really do not understand why you do not want to do this: it is a very
useful low-hanging fruit, and it is quite elegant and fun to implement.
Alas, you did not reply to my last round of arguments.

> + * @param keyvalue_cmp compare function, will be passed the key + value concatenated.

Please no. Pass the key and the value separately to the compare
function.

> +/**
> + * Add a compatible compare function to the map.
> + * The function will later be selected by using AV_MAP_CMP_* flags
> + *
> + * Functions must be added from most specific to least specific, that is if a AVMap is build
> + * with a case insensitive compare no case sensitive compare functions can be added. This is
> + * to avoid building non functional AVMaps.
> + *
> + *
> + * @see av_map_new
> + *
> + * @param cmp compare function to be added.
> + *            cmp(a,b) must return 0 or be equal to the previously added compare function for (a,b), if it returns 0 it also must do so for all
> + *            elements between a and b
> + *
> + * @param cmp_flags a combination of AV_MAP_CMP_*, note key/keyvalue and truncated vs non truncated
> + *                  are mandatory to be specified
> + *
> + * @return a negative error code if the cmp_flags are illegal or unsupported for this AVMap
> + *         If you know your flags are valid, then you dont need to check the return
> + */
> +int av_map_add_cmp_func(AVMap *m, AVMapCompareFunc cmp, int cmp_flags);

This whole thing with multiple compare functions makes the code quite
hard to understand. And not just the code: it makes the API itself hard
to understand.

I think the practical goal it server could be achieved in a simpler way.

If I understand correctly, it servers to allow the equivalent of
AV_DICT_IGNORE_SUFFIX and such. It would be simpler to have a flag to
av_map_find() to let it return the first key >= to the searched key.

Regards,

-- 
  Nicolas George
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

  reply	other threads:[~2025-04-17 17:57 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-17 16:52 Michael Niedermayer
2025-04-17 17:57 ` Nicolas George [this message]
2025-04-17 20:12   ` Michael Niedermayer
2025-04-18 14:46     ` Leo Izen
2025-04-19  1:23       ` Michael Niedermayer
2025-04-19 13:03     ` Nicolas George
2025-04-19 16:23       ` Michael Niedermayer
2025-04-19 17:23         ` Nicolas George
2025-04-19 17:36           ` Nicolas George
2025-04-19 10:36 ` Leo Izen
2025-04-19 14:39   ` Michael Niedermayer

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=aAFBGsbuU5SG2ma4@phare.normalesup.org \
    --to=george@nsup.org \
    --cc=ffmpeg-devel@ffmpeg.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
		ffmpegdev@gitmailbox.com
	public-inbox-index ffmpegdev

Example config snippet for mirrors.


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git